Blade Heat Treatment Myths Busted!

As well as debunking some of the myths out there about heat treating blades, I thought it would be useful to cover some of the principles, so that we’re on the
same page to start.

The four key processes in blade heat treatment are:

  • stress relieving
  • annealing
  • normalising
  • hardening and tempering

Stress Relief and How to Avoid a Bent Knife!

Stress relieving is a procedure to remove induced stresses from blade material (which could result in a bent knife) without changing the microstructure.


This can happen because some blade material is manufactured and shipped as a coil. Before purchase the material is de-coiled (straightened and flattened) so that we can buy a flat (or sometimes not so flat!!!) strip.


This bending and un-bending creates stress and makes the material harder (exactly as bending a paper clip back and forth will harden it until it breaks).


When we subsequently heat up the material for hardening these stresses relieve themselves and the blade often bends.

Now no-one wants a bent knife…so the big question is…how do you stop it happening?

If your heat treatment process uses a system of press quenching this will keep the blade straight, but if you can’t do that then try this…

Bend your blade blank straight and heat it to 600-6500C, (this is a very dull red so remember to check the colour in a low light area). After heating let it air cool and, if necessary, re-straighten it before hardening.

Annealing…(for when you’ve forgotten to drill the hole!)

Annealing is ideal for when you need to rework the material again, maybe to drill a hole.

This involves heating the steel to the hardening temperature and cooling it slowly enough, in the furnace or in vermiculite, to prevent it hardening. For stainless steels it may be necessary to have a furnace that can cool at a very slow (programmed) rate, say 10ºC or 20ºC per hr.


For Carbon Steels simply burying the piece in vermiculite (available from builders’ merchants) is sufficient. Stainless Steels however are air hardening and they’ll still harden somewhat. The easiest solution is to stress relieve as described above. The hardness will still be high but drilling will be possible with care.


Normalising is the process used to ‘normalise’ a deformed microstructure, e.g. after forging. Normalising is similar to annealing except that after heating the steel is usually air cooled and it doesn’t matter if it hardens slightly (or even fully).

Here you’re trying to “normalise” the structure of the steel, not soften it. All steels can be normalised but if any machining is required afterwards then normalising may be the wrong treatment and it’s likely annealing is required.

If you forge blades normalising is not optional but mandatory, because you must correct the microstructural deformation caused by heating the steel to forging temperatures.

So what’s involved in normalising?

To normalise any steel heat it to 30ºc above the hardening temperature (if a hardening range is given use 30ºc above the top of that range) and let it air cool to room temp. Time at temperature need only be the same as for hardening. It is important that it then be allowed to cool to the point you can handle it. (If you’re normalising pattern welded steel use the higher of the temperatures as may be recommended for any of the different constituents.)

Myth #1 – You Need to Normalise Multiple Times!

There is NO advantage in doing this more than once – it works 100% first time providing you’re at the correct temperature. Did you hear that? You only need to Normalise once.

I often hear “I normalise 3 (or 5 or 9) times so that I’ll definitely get it right on one of them. Please believe me – the last one negates all that go before it.

“I normalise several times to ensure I get a fine grain size” – grain size is determined by the temperature of the last normalising cycle – multiple cycles are used to spherodise carbides and we don’t need to do that.

Let’s move on to Hardening and Tempering!

These two processes are treated as one because you should never harden without tempering.

Hardening means to heat the steel to the recommended hardening temperature and cooling it fast enough to force the microstructure to change to a very hard (martensitic) state. This required cooling rate can be as fast as iced brine for very low carbon steels to still air for high alloy stainless steels.

‘Recommended hardening temperature’ means that dictated by the material specification or the steel manufacturer. You cannot improve on this so please don’t try!!!

Steels have a hardening range, usually over 30-400C. The lower temperatures are for thinner sections and the upper for thicker so work near the lower end of the range. If hardening from the forge try to ensure you have a very even colour.

Time at temperature is important but only to ensure that the temperature is even throughout the blade. This table is from Sandvik with respect to stainless knives heat treated in a furnace. For carbon steel blades add 20% to the times as heat transfer rates at the lower temperatures are a bit slower. Start timing when the furnace first recovers to the set point after loading the knife.

Quench rate will also be recommended – quench too fast and you will probably crack the blade. Too slow and it won’t harden properly.

Quench vertically into oil and hold still as it reduces the likelihood of distortion. If you buy pukka quench oil use a slow oil for stainless and medium or fast oil for carbon. If trying to produce a ‘Hamon’ the faster the quench the better the Hamon (but don’t use water!!!)

Whilst on the subject of oil, many use old engine oil and cooking oil. I don’t recommend the former at all. However, when I was a metallurgical apprentice (50+ yrs ago!) we often used Rapeseed Oil. It wasn’t great but this was before the time of precision formulated quenchants. It has a high flash point and will quench thin sections quite well. If you don’t want to pay for quench oil this is the best alternative.

Tempering is another non-optional process. Once hardened the blade is very highly stressed and if used as such will suffer edge chipping at best to outright breakage at worst. All blades must be tempered (and possibly require multiple tempers and/or sub-zero treatments) and again follow the steel manufacturer’s guidelines.

Tempering must follow as soon as practical and a domestic fan oven at 200 works well (don’t get caught if using the one in the kitchen!!!).

For tempering with a flame, firstly grind one face so you have a clean surface free of scale and heat from the other side to a pale or medium straw colour.

Decarburisation (Decarb) – What is it and how it affects your blade…

Decarburisation is a topic that seems to receive little attention and appears to be little understood. A simple definition is the removal of carbon from the surface layer of a piece of steel. If the carbon is removed the steel won’t harden as carbon is the primary alloying element that turns iron into steel.

When we heat our piece of steel in air the carbon is burnt from the surface as it reacts with oxygen in the atmosphere. This leaves a soft surface layer after hardening. The hotter the steel and the longer it’s in contact with air the more the carbon is removed. I’ve seen decarb ranging from a few microns to over 1mm deep.


Decarb can be reduced or prevented by using any of the following methods:-

1. Heat the blade quickly and quench as soon as it is at the hardening temperature to minimise decarburisation
2. Place a protective atmosphere around the blade to stop air from reaching the surface
3. Heat treat in air and make sure enough metal is ground off afterwards to remove the decarburised layer

So how can you make use of these three options if you’re heat treating at home?

Fast heating and quenching:

Ensure the heat treatment furnace is hot (I’m referring to an enclosed refractory lined, temperature controlled home heat treatment oven here, not a gas fired forge but the principle is the same – make sure the forge is hot and stable).

This may sound obvious but consider this. When the furnace is heating up 100% of the power available is used the heat up the refractory lining of the furnace. When it gets to temperature for the first time this is still the case. As heat soaks further into the refractory, the furnace starts to cycle on & off until an equilibrium is reached and the furnace elements may be on for only for 50 – 80% of the time depending how well insulated your furnace is. This spare power is what we now need to heat up our blades quickly.

Therefore, don’t load your first blades into the furnace as soon as it reaches temperature (and do not load blades into a cold furnace) but let it cycle a few times first. I wait for about 15mins once temperature is reached. Don’t put a large load in – 1-2 fixed blades or 3-4 folders at a time. Then start timing your soak period once the furnace has recovered and use the Sandvik recommended soak times – they work well.

Protective Atmosphere:

Placing a protective atmosphere around the blade is partially possible but not, in my opinion, really practical in the home situation. Wrapping the blade in stainless steel foil helps but it rarely prevents decarb, even if the blade is wrapped in paper or paraffin or charcoal is included in the wrap. The air trapped in the foil has to escape as it expands (as does the gasses produced as the paper or paraffin burns) and this escape hole allows oxygen back in through the hole by the process of ‘back diffusion’.

It also requires a longer soak time and this is especially bad for stainless steels (it stabilises retained austenite, which lowers hardness, and lower cryogenic treatments are required). Wrapping is great for the prevention of heat treatment scale but you’ll still get some decarb.

Grinding it off:

If you’re heat treating and cooling quickly as I’ve suggested you won’t get more than a few microns of decarb and as long as you grind this away you won’t have a problem. I do not pre-grind any of my blades prior to the heat treatment for this very reason.

What is important to remember here is that minor decarburisation is not necessarily bad. We just need to understand it and know how to deal with it.

Let’s Talk Temperature

Firstly, I want to talk about temperature of steel and how to measure and/or assess it.

The best option is to have a pukka heat treatment oven with a thermocouple and controller so you know that you’re at the right temperature. However, I appreciate that this is a luxury that you might not have.

All that follows does talk about temperature as if you do have a furnace available, but I’m also going to share how to estimate temperature.

There are two indicators of temperature for iron & Steel, colour and magnetism. Iron & steel are kind in that they’ll tell you how hot they are by their colour. With practice, and much practice is necessary, you can tell the temperature to within a few tens of degrees. Google ‘steel tempering colours’ and you’ll find charts like the one below that have colour indications from 200ºc to 1300ºc.


Below red heat a freshly ground surface and a keen eye is all that’s required. However, once the steel starts to glow red try to be in a low light area, i.e. workshop lights off or in deep shade if outside. Bright light masks the glow and an underestimation of temperature is likely.

Magnetism is also a useful indicator and can be used in conjunction with colour. Steel loses its magnetism at the Curie Point, 768-770ºC (depending which textbook you read!!). This temperature is fixed, it doesn’t vary from alloy to alloy. All alloys have their own specific hardening temperature and all are above 770ºC, some as much as 300ºC higher.

Myth #2 – When steel loses its magnetism it’s at the right temperature for hardening

In our knife making world the recommended hardening temperature can vary from 780ºc to 1100ºc so you can see that magnetism is only a guide that you’re getting there, not a firm indicator that you’re at the correct hardening temperature.

Myth #3 – That heat treating Stainless Steel is NOT the same as for carbon steel

What follows applies equally to Carbon & Stainless steels. However, a word of warning. Stainless Steels are extremely sensitive to hardening temperature. Literally, a 20ºc – 30ºc variation from the ideal hardening temperature can severely compromise the properties obtained.

Some Stainless Steels even give a recommended range of up to 40ºc but where you choose to be within this range being dependant on any sub-zero treatments you have available. In short, Stainless steels are sensitive and best heat treated using accurate (i.e. better than the Mk 1 eyeball) temperature control.

Myth #4 – That cryogenics (deep freezing in dry ice or liquid nitrogen) is essential for Stainless Steel heat treatment.

Sub-Zero treatments are for Stainless Steel blades only. There’s no substantial benefit to carbon steel knives. They are intended to remove retained austenite and if trying to achieve absolute maximum hardness one needs to harden from the very highest recommended temp, and freeze to -90ºC or below to remove most of the retained austenite. However, this will only give an extra 1-2 points Rockwell hardness as well as increasing the brittleness of the knife.

I always use a low to mid-range hardening temp for stainless steel, freeze for 1hr at -25ºC (that’s as low as my freezer will go) and then temper. Don’t temper, freeze and re-temper. The first temper will stabilise the retained austenite by about -50ºC to -90ºC which then negates any benefit from subsequent freezing. Stainless steels should always be tempered twice and freezing between tempers does have some theoretical justification although I doubt any measurable effect.

Additionally, some stainless and tool steels, (D2 is a classic example), will benefit from a re-temper after 5 days as some retained austenite will have degenerated into martensite. The amount is minor however and won’t affect the knife’s performance.

Well that’s my broad take on heat treating. If you have any specific queries please feel free to mail me at